Tests of Relations Flashcards
Which one of these questions would we not use a test of relations to find out?
1. What is the relationship between cups of coffee consumed per day and test scores?
2. Are test scores higher when more coffee is consumed?
3. Is there a difference in test scores between the group that no coffee per day vs the group that had 2 cups per day?
4. Can coffee consumed be used to predict test scores?
3
A correlation does not allow for statements on:
cause and effect
what are the 2 mechanisms for interpretation regarding correlation?
scatter plots with correlation line
correlation coefficient (r)
The higher correlation and correlation coefficients, the closer the data points to the:
line of best fit
As one variable increases, the other variable increases. This is a __________ relationship.
direct
as one variable decreases, the other variable increases. This is a ___________ relationship.
indirect
Test of relations for ordinal and ordinal?
spearmans rank
test of relations for continuous and continuous?
pearsons
test of relations for continuous and categorical?
point biserial
test of relations for categorical and categorical?
phi correlation
what do we use to interpet a correlation coefficient?
portney and watkins
what is the portney and watkins correlation coefficient scale?
little to no relationship - 0 - .25
fair - .25 - .5
moderate -.5 - .75
strong -.75 - 1.00
coefficient precautions include:
- it is a measure of linear relationship only
- interpretation is subjective
- outliers can impact results
- does not give any insight into comparison or causation
what are the 2 mechanisms for interpretation of a regression?
scatter plots with regression line
coefficient of determination (R2)
What is used, a linear regression or a logistic regression, when Y is continuous?
linear regression
What are the 3 major goals of linear regression statistical analysis?
- Calculate R2 to analyze how confidently observed differences in Y can be explained by a difference X.
- Calculate the P value to determine if the R2 value is statistically significant
- Use the line to predict the value of one variable from another.
What is the difference between the correlation coefficient and the coefficient of determination?
R = correlation coefficient
R2 = coefficient of determination
What represents the percentage of the total variance in the Y scores that can be explained by the X scores?
coefficient of determination.
1 - R2 is the percentage of:
the total variance in Y not explained by the X scores.